The identification of novel neuropeptides that act as the endogenous ligand for orphan G Protein Coupled Receptors (GPCRs) has been a slow and costly process that required the purification of an unstable peptide from a brain homogenate. This process can be improved upon using newly developed computational methodologies. A neuropeptide prohormone profile based upon species comparison has been produced. This profile includes a signal sequence, a region of dissimilarity between the species, a splicing site (double basic residues), a region of similarity between the species, and another splicing site. This profile can be modeled computationally using newly developed Match Profile Hidden Markov Model (MPHMM) methodology. The MPHMM computational profile will be used to identify putative novel prohormones and hormones from a dataset of all known and proposed or hypothetical proteins obtained from either the Celera Discovery System (CDS) or from public sources (GenBank). Software tools will be improved upon to enable discriminative training of match modules and to control ranking via visualization of matches. Once putative peptide hormones are proposed, their transcripts will be identified by quantitative rtPCR and localized using in situ hybridization. To identify which of these might be involved in drug abuse, transcript levels will be determined in brain after acute and chronic treatment of mice with morphine, cocaine, and nicotine or after withdrawal from the drug. Any transcript that changes subsequent to treatment with drugs of abuse will be highest priority in the ensuing purification and identification of the proposed neuropeptide. These studies should lead to the identification of novel neuropeptides that could be involved in the drug abuse process.